Return
Optimizing ambiguous speech emotion recognition through spatial–temporal parallel network with label correction strategy
DOI:10.1016/j.cviu.2025.104483.png)
Abstract
En 中文
• An effective label correction strategy is proposed. • Provide a reference method for exploring and processing ambiguous speech. • Offer a new attempt at spatial–temporal information fusion.
Keywords:
label correction
ambiguous speech
spatial-temporal information fusion
Journal
IF:
3.5
Papers:
428
Citations:
7.3K

